Purpose of the study: The objective of this study was to compare the dynamic hip screw (Synthes) and intramedullary fixation (Targon PF, Aesculap) for the treatment of pertrochanteric fractures in terms of stability, complications and cost effectiveness.
Material and methods: This prospective randomized study included 60 patients hospitalized in an emergency setting for pertrochanteric fractures between December 2003 and June 2004. All surgeons in our unit participated in the study. Two fixation systems were used: the Targon proximal femoral nail (Aesculap) and the screw-plate dynamic hip screw (Synthes). We noted: patient status (ASA classification), operative data (type of implant, duration), postoperative data (blood loss, radiographic findings, early complications), and outcome (Harris score, time to walking, mortality). All patients were assessed three months after surgery. The series included 60 patients, 14 men (23%) and 46 women (77%). The Targon PF nail was used for 34 patients and the DHS for 26. In the intramedullary fixation group mean age was 81 years (SD = 12.8, range 23-96); it was 82 years (SD 9.8, range 47-97) in the screw-plate group. The AO classification was: intramedullary fixation 31A1 (n = 11), 31A2 (n = 20), 31A3 (n = 3); screw-plate 31A1 (n = 14), 31A2 (n = 11), 31A3 (n = 1). Thirty-one fractures were stable, 29 unstable. Unstable fractures were treated with the Targon PF nail (n = 18) and the DHS screw-plate (n = 11). Stable fractures were treated with the Targon PF nail (n = 15) and the DHS screw-plate (n = 15).
Results: Mean operative time was 35 minutes for intramedullary nailing and 42 mintues for screw-plate fixation. Mean blood loss was 410 ml for intramedullary nailing and 325 ml for screw-plate fixation (p = 0.07). Mean hospital stay was the same (11 days) in both groups. At three months, mechanical complications involved migration of the cervical screw outside the femoral head for three Targon PF fixations and for two DHS fixations. Screw migration was favored by fracture instability and presence of osteoporosis. There were two deaths in the intramedullary nailing group and one in the screw-plate group. Time to walking was 20 days on average in the intramedullary nailing group and 25 days in the screw-plate group. The mean Harris hip score was 60 in the intramedullary group and 59 in the screw-plate group.
Discussion and conclusion: Data in the literature report an advantage for intramedullary nailing, particularly a mechanical advantage, for the treatment of pertrochanteric fractures. Our findings show that good results are obtained with the screw-plate fixation using the DHS with less blood loss and at a lower cost.